What is being overlooked is a growth segment that is on the rise; Chromebooks (and new Chromeboxes). An ARM chip in a Mac hardware shell need not necessarily run OS X. It can easily just run iOS.
When Mac OS transitioned to PowerPC (from 68K) and to x86 ( from PowerPC), Apple did not already have a
larger OS ecosystem on those targeted processors.
This is also
vastly different than the Windows RT situation.
Actually Apple is in pretty good shape here. MS's had (has) problems that Apple wouldn't:
- MS couldn't / didn't want to port the massive, crusty, and incredibly antiquated Win32 API and the stuff that has grown up around it to ARM. Apple's OS X APIs on the other hand, are well suited to to ARM, especially their own 64-bit chip.
Apple doesn't have to 'port' anything to ARM/iOS laptop. That is the huge difference between what Apple could be doing with an "iOSbook".
- MS also switched from a pointer-based UI to a touch-optimized one which broke compatibility with everything that came before. But the rumor here seems to suggest a desktop UI so no need for Apple to do the same.
Again shocker ..... Apple
already has an established touch based GUI. The most cost effective way of rolling out a touch based laptop would be to just use the extremely highly adopted touch based GUI they already have.
This stark contrast to "Windows" phone/embedded OS
MS also didn't switch GUIs. It was far more the case that both GUIs are present. On Windows RT there just was not any 3rd party apps for the legacy mode "half" of the GUI.
It may be mostly a recompile and testing for most software with only a small set of changes needed in the end... Like the switch from 32-bit to 64-bit for Coco-based apps. Where apps will get stuck is if Apple drops support for old or obscure APIs like with Carbon.
Actually one of the big fails of RT is not having an emulation mode. Neither of Apple's very successful skipped bundling a transparent emulation mode. NeXT didn't do one and was not particularly successful. Asking the entire ecosystem to recompile and ship 'fat binaries' isn't a solution that is going to avoid short term problems.
Additionally with Apple's network only software distribution methodology 'fat binaries' is a also a problem because it means large bulk.
Apple will go all ARM, or no ARM. No pick and mixing like Microsoft.
There is no good reason to go all ARM just for the sake of being all ARM. ARM based solution aren't equivalent to the Intel offerings. That whole A7 is "desktop class" is stretched so far out of context of fullfilling broad spectrum Mac hardware needs it is more humorous than factual. It is just seeds to stir the pot of a long discussion thread ( to feed Macrumors money with ad views).
The 64 bit ARM gets into the Intel ATOM (and low end AMD) range of performance. Core i5/i7 still run circles around 64 bit ARM in terms of performance.
A total conversion with the Apple developer environment would force devs to adapt just as they did with the intel transition, and that was pretty clean all things considered.
It was clean because devs could transition via the emulator and fat binaries. Both of those are more problematical now. Apple didn't even do last emulator they used ( so internal talent is questionable) and bulky binaries are going to force even more folks at the end of slower Internet pipes to look for other solutions.
Devs that work on both OS X and iOS apps have been dealing with Intel and ARM happily for the past 7 years.
I suspect that they'd market this as something that is a third way.
Why do they need a 3rd way? Why wouldn't iOS apps work? If aiming into the class iOS device price zone $200-900 then Apple could target that with iOS devices. That still leaves the Macs mainly in the $900+ ( the Mac Mini the 'odd' exception).
There a fair number of folks buying Chromebooks/Chromeboxes. The folks who just need mostly a browser, email , video chat , pictures , music, and photos ( similar to what the iPad was aimed at initially) coupled to a keyboard and trackpad could very effectively accomplish all that with just iOS apps. In fact, there would be more apps diversity than what Chrome devices have.
While the Ax SoCs are powerful, they won't displace the Mac Pro any day soon.
So yes... why fragment the entry level Mac laptop line up from the rest of the Mac line up???? Especially when there is no deep seated need right now.
Where the Mac line should be aimed at is taking classic (from 4-8 years ago) Mac Pro workload and moving it down to the rest of the Mac line up. ARM (even the current 64 bit variant ) isn't particularly up to that task. Core i5/i7 is.
I hope Apple does not do Microsoft's mistake; the whole 'Windows RT' mess was for that reason, as Surface RT and Asus RT tablets both have ARM processors.
A year later Intel releases new lower power lower heat ATOM processors that literally render the whole thing useless
One of RT's major problem was far more not having an established application ecosystem. Apple doesn't have the problem of how to roll out a large ARM based app ecosystem.
Another task RT was saddled with was trying to get developers to develop for a new GUI where touch was at least just as important as the mouse pointer. Again Apple already has accomplished that over the last 7 years.
Part of what RT was trying to do is merge Windows Phone with 'Windows' so as to pull the Windows Phone into greater adoption. Apple doesn't have an iiOS or OS X adoption problem.
Apple probably does have an increasing "too expensive systems" problem. Much of the pinning for a ARM "Mac" is really for a less expensive Mac. If Apple went A7+/A8 ARM mac and charged current Mac prices I suspect you would find most of the ARM cheerleaders in this thread evaporate as being vocal for that change.
The real core issue is that a large segment of users workload needs are plateauing. It isn't that the A7 is "just as fast" as the desktop offerings from Intel/AMD as much as the A7 is "fast enough" for their everyday workload. The shift is more so about "more affordable" than "faster".
The notion that Apple has to collapse down to just one OS is highly questionable. iOS and OS X share a common subset so it is not all that expensive to do both as long as both iOS device and Mac ecosystems are growing and/or quite profitable.
Apple helped pushed Intel toward better iGPU and lower power and Intel is delivering. Until Intel completely drops the ball or the x86 market in a direction different from where the Mac ecosystem is going long term it makes little to no sense to shift to ARM just because of some "higher homogenous" stance. In fact, making iOS devices more homogenous with Mac devices would be as much as hindrance to Mac as helpful. If largely the exact same hardware that makes differentiation all that much harder.
